Dr Schetelig on ATLG vs PTCY as GVHD Prophylaxis in Hematologic Malignancies

What was a surprise was that we saw very low non-relapse mortality related to less infectious death with the use of ATLG, which translated into lower non-relapse mortality and even gave a signal of potentially improved overall survival.

Johannes Schetelig, MD, MSc, head of the Stem Cell Transplantation Unit at University Hospital Carl Gustav Carus Dresden and director of clinical research at DKMS, discussed findings from the randomized, noninferiority phase 3 GRAPPA trial (NCT05153226) comparing post-transplant cyclophosphamide (PTCy) with anti–T-lymphocyte globulin (ATLG) as graft-vs-host disease (GVHD) prophylaxis in patients undergoing matched unrelated donor allogeneic hematopoietic stem cell transplant (allo-HSCT).

The trial, presented at the 2026 EHA Congress, was designed to determine whether PTCy could be considered noninferior to ATLG for overall survival (OS) using a 5% margin, with GVHD/relapse-free survival (GRFS) serving as a co-primary end point.

As Schetelig and colleagues anticipated, PTCy more potently suppressed grade 2 to 4 acute GVHD (adjusted HR, 0.70; 95% CI, 0.51-0.96) and chronic GVHD of any grade (adjusted HR, 0.59; 95% CI, 0.45.077), including moderate-to-severe disease (adjusted HR, 0.71; 95% CI, 0.41-1.23). However, that reduction in GVHD did not translate into a survival benefit, Schetelig explained. There was no difference in the risk of relapse between the arms (adjusted HR, 0.86; 95% CI, 0.63-1.16), but PTCy was associated with a higher risk of non-relapse mortality (NRM; HR, 1.86; 95% CI, 1.09-3.17; P = .02). This increase in NRM was driven by infectious deaths among patients who did not develop GVHD, according to Schetelig.

In the intention-to-treat population, at a median follow-up of 2.3 years, the 2-year OS rate was 75% (95% CI, 69%-80%) with ATLG (n = 275) vs 68% (95% CI, 63%-73%) with PTCy (n = 383; adjusted HR, 1.341; 95% CI, 0.999-1.801; P = .85), and the test for superiority with ATLG30 bordered on statistical significance (P = .051). The 2-year NRM rate was 7% (95% CI, 4%-11%) with ATLG vs 13% (95% CI, 9%-16%) with PTCy.

Since noninferiority of PTCy could not be demonstrated, ATLG remains the standard of care for patients with matched unrelated donors, Schetelig said, noting that ATLG is not approved in all regions, including the United States. For regions with ATLG approved, the trial’s findings may help inform efforts to improve immune reconstitution after PTCy. Ongoing trials are evaluating lower cyclophosphamide doses to shorten the period of neutropenia and enhance immune reconstitution, as well as reduced post-transplant immunosuppression, he added. In a future randomized controlled trial, ATLG will again serve as the control arm, with PTCy most likely among the experimental options, Schetelig concluded.